Pre-conference workshop: Visual mismatch negativity
Tuesday, Sep 8, 2015
Visual mismatch negativity, prediction, and its functional roles
1National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba, Japan
2University of Zurich, Zurich, Switzerland
Visual mismatch negativity (MMN) is an event-related brain potential (ERP) component that is automatically elicited when sequential rules inherent in a visual stimulus sequence are violated. So far, visual MMN has been observed in response to repetition-rule-violating stimuli (i.e., deviant stimuli) that are occasionally inserted in a sequence of repetition-rule-conforming stimuli (standard stimuli in an oddball sequence), as well as change-rule-violating stimuli (irregular stimuli) that are occasionally inserted in a more complex sequence of change-rule-conforming stimuli (regular stimuli). Based on these previous findings, recent theories have proposed that visual MMN is associated with stimulus-context-based prediction; that is, (1) sequential rules embedded in the temporal context or temporal structure of a visual stimulus are extracted, (2) a predictive model encoding the extracted sequential rules is established, (3) predictions about the forthcoming visual event are formed based on the predictive model, and (4) representations of the current and the predicted visual events are compared. When incongruence between them has been detected, visual MMN is elicited (Kimura, Schröger, & Czigler, 2011, NeuroReport; Kimura, 2012, International Journal of Psychophysiology). Furthermore, according to the hierarchical predictive coding framework, it has been proposed that visual MMN reflects a perceptual prediction error signal in the brain, which plays a functional role in updating the predictive model (Stefanics, Kremláček, & Czigler, 2014, Frontiers in Human Neuroscience). We will discuss the essence of these theoretical proposals and the possible future directions for more precise understanding of the functional role of visual MMN.